A CMOS image sensor has been known as a solid-state imaging element (an image sensor). The CMOS image sensor has been used for a digital still camera, a digital video camera, and various mobile terminal devices, such as a mobile phone with a camera.
In most cases of image sensors in the related art, a signal is extracted in such a manner that a charge of the photodiode (PD) is transferred to a floating diffusion (FD), and then potential of FD is subjected to sensing. In this case, when capacity of FD is small, gain converted into voltage increases, and thus subsequent noise is reduced. Therefore, satisfactory S/N is ensured and an amount of usable charges is small. When capacity of FD is large, the amount of usable charges is large and S/N is small.
A configuration in which FD is constituted by two capacitors to solve the trade-off between the amount of charges and S/N, and thus, when the amount of charges is small, a signal (S1) sensed by one capacitor is used and, when the amount of the charges is large, a signal (S2) sensed by two capacitors is used has been proposed (for example, see PTLs 1 to 3).
Even in a case to which the technology of this proposal is applied, an actual signal is obtained by calculating a difference between S1 or S2 and one or two reset level signals (N) in a state where FD1 or FD2 is reset, similarly to a case in which one FD is provided.